clustering::hdbscan::BoruvkaMstBackend< T > Class Template Reference

KDTree-accelerated Boruvka MST backend over mutual-reachability distances. More...

#include <clustering/hdbscan/policy/boruvka_mst_backend.h>

Public Member Functions
	BoruvkaMstBackend ()=default
void	run (const NDArray< T, 2 > &X, std::size_t minSamples, math::Pool pool, MstOutput< T > &out)
	Build the MRD-weighted minimum spanning tree of X.

Detailed Description

template<class T>
class clustering::hdbscan::BoruvkaMstBackend< T >

KDTree-accelerated Boruvka MST backend over mutual-reachability distances.

Given a point matrix X and core-distance parameter minSamples, the backend:

1. Builds a KDTree on X.
2. Computes per-point core distances as the minSamples-th nearest-neighbour squared distance (self-excluded) via KDTree::knnQuery.
3. Runs Boruvka rounds: each round invokes clustering::hdbscan::detail::nearestOutComponent to find the minimum-MRD-weight out-of-component edge per component via per-point KDTree traversals pruned on each subtree's axis-aligned bounding box, then unions those edges into the spanning tree under the union-find component structure.
4. Sorts the resulting edges ascending by weight for a canonical output ordering.

The backend satisfies MstBackendStrategy and holds no persistent state across calls; data-dependent state (the KDTree) is built fresh each run in line with the HDBSCAN pipeline contract that forbids cross-call caching of pointer-keyed data.

Template Parameters

T	Element type of the point cloud. Only float is supported; a double specialization is out of scope in line with HDBSCAN's top-level restriction.

Definition at line 41 of file boruvka_mst_backend.h.

Constructor & Destructor Documentation

BoruvkaMstBackend()

template<class T>

clustering::hdbscan::BoruvkaMstBackend< T >::BoruvkaMstBackend ( )

default

Member Function Documentation

run()

template<class T>

void clustering::hdbscan::BoruvkaMstBackend< T >::run ( const NDArray< T, 2 > & X, std::size_t minSamples, math::Pool pool, MstOutput< T > & out )

inline

Build the MRD-weighted minimum spanning tree of X.

Precondition

minSamples is positive and strictly less than X.dim(0).

X.dim(0) fits in std::int32_t; the HDBSCAN class shell asserts this before a call reaches the backend.

Parameters

X	Contiguous (n x d) dataset; caller retains ownership.
minSamples	Neighbour count driving the core-distance definition.
pool	Worker pool; forwarded to the kNN query and the per-round KDTree traversal. A null pool runs single-threaded.
out	Destination; edges filled with n - 1 entries in ascending-weight order and coreDistances sized to n.

Definition at line 63 of file boruvka_mst_backend.h.

---

The documentation for this class was generated from the following file:

• include/clustering/hdbscan/policy/boruvka_mst_backend.h